Predicting Ruthenium Catalysed Hydrogenation of Esters using Machine Learning

Catalytic hydrogenation of esters is a sustainable approach for the production of fine chemicals, and pharmaceutical drugs. However, the efficiency and cost of catalysts are often the bottlenecks in the commercialization of such technologies. The conventional approach of catalyst discovery is based on empiricism that makes the discovery process time-consuming and expensive. There is an urgent need to develop effective approaches to discover efficient catalysts for hydrogenation reactions. We demonstrate here the approach of machine learning for the prediction of out-comes for the catalytic hydrogenation of esters. Our models can predict the reaction yields with high mean accuracies of up to 91% (test set) and suggest that the use of certain chemical descriptors selectively can result in a more accurate model. Furthermore, cata-lysts and some of their corresponding descriptors can also be pre-dicted with mean accuracies of 85%, and >90%, respectively.

Comprehensive analysis of commercial fragment libraries

Screening of fragment libraries is a valuable approach to the drug discovery process. The quality of the library is one of the keys to success, and more particularly the design...

2.5 CuAAC and Metal-Free 1,3-Dipolar Huisgen Cycloadditions in Drug Discovery

Proclaimed by Sharpless in 2001, the manifesto of click chemistry philosophy shifted the focus from target-oriented to drug-like-oriented synthesis, and has enormously accelerated the drug-discovery process over the last two decades. Copper(I)-catalyzed and metal-free versions of the Huisgen 1,3-dipolar cycloaddition of azides and alkynes have become the reference click chemistry synthetic tools. These processes are adaptable to various drug-design modes such as kinetic target guided synthesis (in situ click chemistry assembling; KTGS), combinatorial chemistry/high-throughput-screening approaches, or structure-based rational projecting. Moreover, the facile click chemistry derivatization of natural or synthetic products, linking molecules or improving the stability of leads by installation of 1,2,3-triazoles, is another important stream of bioactivities. This review is intended to provide a general overview of click-chemistry-powered drug design, with dozens of successful examples resulting in the discovery of nanomolar-active 1,2,3-triazoles in every stage of drug development.

Human disease biomarkers: challenges, advances, and trends in their validation

Biomarkers are important tools in the medical field, once they allow better prediction, characterization, and treatment of diseases. In this scenario, it is essential that biomarkers are highly accurate. Thus, biomarker validation is an essential part of ensuring the effectiveness of a biomarker. Validation of biomarkers is the process by which biomarkers are evaluated for accuracy and consistency, as well as their ability to inform the condition of health or disease. Although, there is no unique measure that can be used to determine the validity for all biomarkers, there are general criteria that all biomarkers must meet to be useful. In this work, we review the definition of biomarkers and discuss the validity components. We then critically discuss the main methods used to validate biomarkers and consider some examples of biomarkers of the diseases which most killer in the world (cardiovascular diseases, cancer, and viral infections), highlighting the potential biochemical pathways of these biomarkers in the biological system. In addition, we also comment on the omic strategies used in the biomarker discovery process and conclude with information about perspectives in biomarker validation through imaging techniques.

The iterative application of a large chemical space in the drug discovery process

Aim. To demonstrate the advantages of large-scale virtual libraries generated using chemical protocols previously validated in primary steps of the drug discovery process.Results and discussion. Two validated parallel chemistry protocols reported earlier were used to create the chemical space. It was then sampled based on diversity metric, and the sample was subjected to the virtual screening on BRD4 target. Hits of virtual screening were synthesized and tested in the thermal shift assay.Experimental part. The chemical space was generated using commercially available building blocks and synthetic protocols suitable for parallel chemistry and previously reported. After narrowing it down, using MedChem filters, the resulting sub-space was clustered based on diversity metrics. Centroids of the clusters were put to the virtual screening against the BRD4 active center. 29 Hits from the docking were synthesized and subjected to the thermal shift assay with BRD4, and 2 compounds showed noticeable dTm.Conclusions. A combination of cheminformatics and molecular docking was applied to find novel potential binders for BRD4 from a large chemical space. The selected set of predicted molecules was synthesized with a 72 % success rate and tested in a thermal shift assay to reveal a 6 % hit rate. The selection can be performed iteratively to fast support of the drug discovery.

A MAC Layer Protocol For Smart Indoor Inventory Management System

The indoor inventory system is gaining more research attention and commercial value with the development of IoT. In this thesis, we presented the design of a MAC protocol that allows synchronized transmission of location and sensing data in a wireless positioning and sensor network for an indoor inventory system. The network supports real-life industrial applications and provides a highly specific positioning method.<div>In the network, mobile sensing tags are connected to smart readers that performs localization of tags and gathers sensing data from the tags. The readers are connected to the back-end cloud. The proposed MAC serves multiple classes of mobile tags with different priorities and latency requirements. These tags transmit critical, position and sensing data with different QoS requirements. The proposed MAC is a hybrid MAC that offers contention-based period for tag discovery and scheduled period for the transmission of sensing data with guaranteed latency. We conducted simulation to evaluate the performance of different methods of discovery process and their impact on latency assurance. We also developed a queuing model to analyze the relationship between parameters, acquiring parameters through experiment, and calculation of boundary values.<br></div><div>Simulation using MatLabTM software suggests that the joining period in design can increase the transmission success rate of high priority messages at the cost of a slight increment in the delay of low priority messages. Preliminary analysis suggests that by adaptively allocating the channel resources of the network to three types of tags, service efficiency can be improved. This result also guides the direction for further improvement.<br></div><div>We explored the performance of two options considered currently, which is selecting the discovery process according to modulo result of unique 16-bit tag ID and random select of an available discovery process. In the current environment where each tag does not have any information about other tags inside the network, the two methods have the same effect on avoiding collisions that could happen in a single discovery cycle.<br></div><div>The proposed MAC layer protocol can provide the best service when the available discovery process in the discovery cycle is for initialization and resetting. For an emergency, the joining period designs can still ensure a success rate for critical messages to be over 90%. Hence, the simulation results indicate the joining period method is able to improve MAC-layer performance.</div><div> <br></div>

A MAC Layer Protocol For Smart Indoor Inventory Management System

The indoor inventory system is gaining more research attention and commercial value with the development of IoT. In this thesis, we presented the design of a MAC protocol that allows synchronized transmission of location and sensing data in a wireless positioning and sensor network for an indoor inventory system. The network supports real-life industrial applications and provides a highly specific positioning method.<div>In the network, mobile sensing tags are connected to smart readers that performs localization of tags and gathers sensing data from the tags. The readers are connected to the back-end cloud. The proposed MAC serves multiple classes of mobile tags with different priorities and latency requirements. These tags transmit critical, position and sensing data with different QoS requirements. The proposed MAC is a hybrid MAC that offers contention-based period for tag discovery and scheduled period for the transmission of sensing data with guaranteed latency. We conducted simulation to evaluate the performance of different methods of discovery process and their impact on latency assurance. We also developed a queuing model to analyze the relationship between parameters, acquiring parameters through experiment, and calculation of boundary values.<br></div><div>Simulation using MatLabTM software suggests that the joining period in design can increase the transmission success rate of high priority messages at the cost of a slight increment in the delay of low priority messages. Preliminary analysis suggests that by adaptively allocating the channel resources of the network to three types of tags, service efficiency can be improved. This result also guides the direction for further improvement.<br></div><div>We explored the performance of two options considered currently, which is selecting the discovery process according to modulo result of unique 16-bit tag ID and random select of an available discovery process. In the current environment where each tag does not have any information about other tags inside the network, the two methods have the same effect on avoiding collisions that could happen in a single discovery cycle.<br></div><div>The proposed MAC layer protocol can provide the best service when the available discovery process in the discovery cycle is for initialization and resetting. For an emergency, the joining period designs can still ensure a success rate for critical messages to be over 90%. Hence, the simulation results indicate the joining period method is able to improve MAC-layer performance.</div><div> <br></div>

Screening of Potential Indonesia Herbal Compounds Based on Multi-Label Classification for 2019 Coronavirus Disease

Coronavirus disease 2019 pandemic spreads rapidly and requires an acceleration in the process of drug discovery. Drug repurposing can help accelerate the drug discovery process by identifying new efficacy for approved drugs, and it is considered an efficient and economical approach. Research in drug repurposing can be done by observing the interactions of drug compounds with protein related to a disease (DTI), then predicting the new drug-target interactions. This study conducted multilabel DTI prediction using the stack autoencoder-deep neural network (SAE-DNN) algorithm. Compound features were extracted using PubChem fingerprint, daylight fingerprint, MACCS fingerprint, and circular fingerprint. The results showed that the SAE-DNN model was able to predict DTI in COVID-19 cases with good performance. The SAE-DNN model with a circular fingerprint dataset produced the best average metrics with an accuracy of 0.831, recall of 0.918, precision of 0.888, and F-measure of 0.89. Herbal compounds prediction results using the SAE-DNN model with the circular, daylight, and PubChem fingerprint dataset resulted in 92, 65, and 79 herbal compounds contained in herbal plants in Indonesia respectively.